|Publication number||US20060103380 A1|
|Application number||US 11/273,259|
|Publication date||May 18, 2006|
|Filing date||Nov 15, 2005|
|Priority date||Dec 11, 2002|
|Also published as||US7254286|
|Publication number||11273259, 273259, US 2006/0103380 A1, US 2006/103380 A1, US 20060103380 A1, US 20060103380A1, US 2006103380 A1, US 2006103380A1, US-A1-20060103380, US-A1-2006103380, US2006/0103380A1, US2006/103380A1, US20060103380 A1, US20060103380A1, US2006103380 A1, US2006103380A1|
|Inventors||Vladimir Kochergin, Philip Swinehart, Mokhtar Maklad|
|Original Assignee||Lake Shore Cryotronics, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Referenced by (11), Classifications (4), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application claims the benefit of provisional application 60/627,907 filed Nov. 16, 2004, incorporated herein by reference. This application also claims the benefit of commonly assigned copending application Ser. No. 10/10/316,192 filed Dec. 11, 2002 of Kochergin et al. entitled “MAGNETO-OPTICAL SENSOR EMPLOYING PHASE-SHIFTED TRANSMISSION BRAGG GRATINGS” (attorney docket 340-70) also incorporated herein by reference.
The technology herein is related to the optical sensing of magnetic fields, including magnetic fields generated by electrical currents. More particularly, the exemplary illustrative non-limiting technology relates to planar waveguide resonator structures, including one or more layers of magneto-optically active material, for the detection of the magnetization changes in magneto-optically active layers by spectral shifts of the characteristic waveguide resonator spectral features, and which are easily connectable to optical fibers.
Optical sensors have significant advantages over electronic sensors in many applications; for instance, those requiring absolute explosion proofing, insensitivity to electromagnetic interference (EMI), low to zero electronic emissions, high voltage exposure during electrical current measurements, and/or complete absence of metals or other electrical conductors. Sensors connected to and/or interrogated through optical fibers (fiber optic sensors; or FO sensors) are being rapidly developed, although the industry is still in its infancy.
One type of FO sensor that has been difficult to reduce to a practical industrial product is a magneto-optical fiber optic sensor (MOFO). This is perhaps partly because of the lack of magnetic sensitivity in most optical materials, lack of the proper form of materials that do have high sensitivity and partly because of the polarization-dependent nature of the sensors that can be made from existing materials.
The illustrative exemplary non-limiting technology herein provides MOFO sensors and methods of manufacture that substantially eliminate the uncertainties in the magnetic field signals due to polarization dependence, provide high signal-to-noise ratios (SNR) and replace the light intensity (amplitude) dependence of prior art MO sensors with signals based on spectral changes in the interrogating light wavelength band.
Prior art MO sensors have been based on the rotation of light polarization during transmission through a material with a high Faraday Effect (e.g., Bi-doped, yttrium-iron garnets, or Bi:YIG) (see, for example Rochford K. B., Rose A. H., Deeter M. N., Day G. W. “Faraday effect current sensor with improved sensitivity-bandwidth product.” Optics Letters, vol. 19, (no.22), November 1994, p. 1903 and Day G. W., Deeter M. N., Rose A. H.; Rochford K. B. “Faraday effect sensors for magnetic field and electric current.” Proceedings of the SPIE—The International Society for Optical Engineering, vol. 2292, (Fiber Optic and Laser Sensors XII, San Diego, Calif., USA, 25-27 July 1994.) 1994, p. 42), a high Verdet constant (rare-earth doped silica glass) (see, e.g., Y. N. Ning et al in “Recent progress in optical current sensing techniques”, Review of Scientific Instruments 66 (5), May 1995, pp. 3097-3111), long lengths of low Verdet constant glass fiber (lengths of a kilometer or more) (see for example [Rose A. H. “Playing with fire and fibers”, IEEE Circuits and Devices Magazine, vol. 15, (no.5), IEEE, September 1999. p. 41]) or reflection from a material with a substantial Kerr effect (ferromagnetic metals) [Oliver, S. A. and C. A. DiMarzio, U.S. Pat. No. 5,736,856; U.S. Pat. No. 5,631,559; U.S. Pat. No. 5,493,220]. In the scheme attributed to Malus, the magnetic field signal from each of these sensors, whether connected by optical fiber or not, is an amplitude signal because crossed polarizers or similar techniques are used to detect the polarization rotation. The light rotated by an external magnetic field measurand is detected as a varying light intensity, or amplitude. The signals are thus vulnerable to amplitude variability and interference due to fiber bending losses, dirty connectors and other similar causes, and large inaccuracies can occur. Polarization-maintaining (PM) fiber can be used to avoid some of the errors due to spurious polarization rotation in the fiber, but cannot solve other amplitude losses. The alternative measurement technique, which employs Sagnac interferometers (J. Blake et al., IEEE Transactions on Power Delivery, Vol. 11, No. 1, January 1996; Moon Fuk Chan; Guansan Chen; Demokan M. S.; Hwa Yaw Tam “Optimal sensing of current based on an extrinsic Sagnac interferometer configuration”, Optics and Lasers in Engineering, vol. 30, (no.1), July 1998. p. 17), relies on phase measurements but is also vulnerable to amplitude and polarization variability, although not at such a high degree as the Malus scheme.
Another difficulty with prior art Faraday Effect sensors, excluding those of the type based on kilometer-length fiber coils, is that the sensors utilize either relatively large blocks of high Verdet constant glass (several centimeters) or small blocks of a high Faraday Effect material such as perpendicular magnetization Bi:YIG. Here, perpendicular magnetization is taken to mean that the magnetization vector in the material is perpendicular to the wafer plane, with no external magnetic field present. This orientation of the magnetization can be arranged if the wafer is cut from a YIG crystal boule, or if a thick YIG film is grown on a substrate wafer, such as is done with liquid phase epitaxy (LPE) on a gadolinium gallium garnet (GGG) substrate, or on an expanded lattice GGG variation doped with materials such as calcium, magnesium, zirconium or scandium. In these cases, the transmission of light is parallel to the magnetization vector and the magnetic field is detected perpendicular to the film plane. This is because the magnetization vector is constrained from rotating due to the shape anisotropy of the material block, preventing any signal from being caused by magnetic fields perpendicular to the light path. In particular with the YIG Faraday Effect sensors, the sensor may be made up of several components (AIRAK, Inc. patent #U.S. Pat. No. 6,534,977) and then glued or otherwise mounted to the fiber, making a fragile and unreliable sensor.
Still further difficulty is found with prior art magnetic field sensors if electrical current is to be measured. With kilometer-length, low Verdet constant fibers, coils are made and the wire is passed through the center. With high voltage power cables, the only way this can be done conveniently is to deliver the sensor with a de-mountable buss bar system—a very costly method. The same is substantially true of high Verdet constant glass sensors, since the blocks of glass must be fashioned into prisms and placed around the buss bar in a robust holding assembly so the optical path circumvents the buss bar—also a very costly method. On the other hand, the YIG sensors such as those disclosed in U.S. Pat. No. 6,534,977 are essentially “point” sensors, but suffer from the need to assemble them from many components that must be kept clean and aligned in their package. This is in addition to the difficulties with amplitude, polarization and sensor robustness discussed above.
The illustrative exemplary non-limiting implementations herein provide sensors based on planar optical waveguide resonators with waveguide modes at least partially localized in the one or more layers of materials exhibiting the Faraday Effect (i.e. magneto-optical materials). The magnetization in such magneto-optical materials is preferably the in-plane type of magnetization, i.e., materials in which the magnetization vector lies in, or nearly in, the plane of the wafer from which the sensor is fashioned. Further, the optical waveguide(s) lies in the plane of the wafer as well.
Exemplary illustrative non-limiting implementations also provide MOFO sensors in which the Faraday Effect material can remain a simple plane-parallel disk shape with no cuts or corners, allowing the magnetization vector to rotate freely in the plane of the sensor disk, greatly increasing the measurement sensitivity and resolution.
Exemplary illustrative non-limiting implementations can provide MOFO sensors with very few parts (the planar chip, the fiber connections and the package) such that the sensor is robust and stable.
Exemplary illustrative non-limiting implementations may also provide MOFO sensors that convert the external field magnitude to an optical wavelength shift that is accurately and unambiguously detectable. Further, crossed polarizers are not required, and the polarization sensitivity is reduced to at least the second order, allowing polarization-maintaining fiber connections to eliminate over-all polarization dependence.
Exemplary illustrative non-limiting implementations further provide effective “point” magneto-optical sensors for the detection of magnetic fields and magnetic field gradients due to electrical currents, permanent magnets, the earth's magnetic field or any other source.
These and other features and advantages provided by exemplary non-limiting illustrative implementations will be better and more completely understood by referring to the following detailed description in connection with the drawings, of which:
An exemplary illustrative non-limiting magneto-optical waveguide sensor is based on MO effects occurring in optical waveguide resonant structures with waveguide modes at least partially localized in the magneto-optical layers, which effects are the so-called nonreciprocal magneto-optic effects of transverse magnetic (TM) waveguide modes [Bahlmann N., Chandrasekhara V., Erdmann A., Gerhardt R., Hertel P., Lehmann R., Salz D., Schroteler F. J., Wallenhorst M., Dotsch H., “Improved design of magnetooptic rib waveguides for optical isolators”, Journal Of Lightwave Technology 16: (5) 818-823 (1998)], [Wallenhorst M., Niemoller M., Dotsch H., Hertel P., Geghardt R., Gather B., “Enhancement of the Nonreciprocal Magnetooptic Effect of TM Modes using Iron-Garnet Double-Layers with Opposite Faraday-Rotation”, JAP, 77: (7) 2902-2905 (1995)]. The physical basis of the sensing mechanism is the dependence of the effective refractive index of the TM mode in the waveguide structure containing a magneto-optical layer in the plane perpendicular to the mode propagation direction projection of the magnetization in the magneto-optical layer. This effect is nonreciprocal, i.e., sign-sensitive. Further, the exemplary illustrative non-limiting implementation provides for the conversion of said magnetization-dependent change in the effective refractive index of the TM waveguide mode into a spectral shift of a resonance spectral feature of the sensor. According to one of the aspects of the exemplary illustrative non-limiting implementation, the said conversion can be performed by fabricating a Bragg grating resonance structure in said waveguide and monitoring the change of the effective refractive index of the TM mode by the resulting change of the spectral position of the Bragg resonance, which is in turn due to changes in the value and/or direction of the external magnetic field. According to a further aspect of the exemplary illustrative non-limiting implementation, a second waveguide resonator structure in the shape of a ring or disk will also convert the magnetic field-induced change of the effective refractive index of the TM waveguide mode into a spectral change of the resonance feature position (i.e., the spectral location of the transmission valley). In all these aspects of the exemplary illustrative non-limiting implementations, the transverse electric (TE) component of the interrogating light is not affected by interaction with the magnetic material and can be used as a reference to separate the magnetic field-induced signal from the spectral shifts caused by temperature, strain or other parameter variations.
Two non-limiting example illustrative non-limiting types of magneto-optically active material are herein disclosed for the fabrication of planar waveguide structures for magnetic field sensing. Each type of material is disclosed in two main non-limiting implementations, grating waveguides and ring resonators. Although many variations in each of these implementations can be made by changes in material composition and device geometry, such alterations in no way limit the scope of the attached claims. Two example divisions of magneto-optically active materials are ferrimagnetic iron garnets and high Verdet constant glasses, which are paramagnetic, but other materials could be used as well.
Sensor design utilizing ferrimagnetic iron garnet material has to take into account the usual magnetic properties familiar to those knowledgeable in the art, such as demagnetization factors, shape and magneto-crystalline anisotropies, saturation magnetization, domain motion, Barkhausen noise, and the like. It is desirable from these points of view that the ferrimagnetic Faraday Effect materials have the magnetization vector in-plane, with the magnetization vector being either free to rotate in the plane of the substrate in response to changing magnetic fields (so-called “easy plane” type of magnetic anisotropy) or with magnetization vector (M) having a preferential and known orientation in the absence of external magnetic fields (so-called in-plane, hard axis type of anisotropy). A relatively thin film of a material, limited to several micrometers thick, is frequently necessary as one requirement to keep the magnetization vector M in the plane of the film, in the absence of the external magnetic field, by means of the demagnetization energy. If the film is too thick, the magnetization vector can swing out-of-plane even in the absence of an external magnetic field.
The iron garnet film of the exemplary illustrative non-limiting implementation is preferably grown epitaxially on a single crystal substrate, which can be matched as closely in crystal lattice constant as possible to that of the film. The MO film can be grown in disk shape or can be formed into a disk shape after growth by any etching means common in the materials processing art, the purpose of which is to avoid the formation of magnetic poles at the film periphery and thus to allow M to rotate freely under the influence of an external magnetic field.
The preferred, but not exclusive, form of a structure allowing free rotation of M is a disk with smooth surfaces and edges as shown in
For the purposes of exemplary illustrative non-limiting implementations, it is most desirable that the Faraday Effect material be a single crystal film in the shape of a disk with smooth surfaces, if the waveguide is to be formed with an applied, non-magnetic transparent dielectric material such as, in a nonlimiting example, silicon dioxide or a polymer. In such a design, the refractive index of the magneto-optical material should be higher than that of the substrate (such as a YIG layer with refractive index in the range of 2.2 to 2.4 on the top of GGG substrate with refractive index of about 1.9) and also higher than that of the non-magnetic transparent dielectric material (for silica, n is about 1.486). There are, however, some alternative aspects of the exemplary illustrative non-limiting implementation that would make use of an applied, non-magnetic waveguide material with a higher index than that of the YIG. The thickness of the Faraday Effect layer may be slightly less than that needed to support a single waveguide mode at the wavelength of the interrogation instrument. The applied rib waveguide structure in a transparent layer on the top of the Faraday Effect layer will help to overcome the waveguide mode cut-off size condition and will permit the guiding of the waveguide mode while keeping said mode localized mostly in the Faraday Effect material to maximize the sensitivity of the sensor to the external magnetic field. In this case, the MO material is also part of the waveguide. Alternatively, in another aspect of the exemplary illustrative non-limiting implementation, the layer exhibiting the Faraday Effect can be used as the substrate, while the waveguide mode can be mostly confined in the “rib” waveguide formed out of thicker, high refractive index material on the top of the Faraday Effect layer. In another aspect of the exemplary illustrative non-limiting implementation, the rib waveguide can be formed in the Faraday Effect layer itself by means of ion milling, reactive ion etching or any other technique well known in the art. It should be noted that it is less desirable that the waveguide be formed in the ferrimagnetic material, since sharp corners and contours will cause extraneous magnetic poles to form and adversely affect the ability of the magnetization vector to rotate freely. However, such a sensor will show some degree of sensitivity.
Another concern that we address is the matching of the optical characteristics of the input and output optical fibers and the sensor waveguides. The numerical apertures (NA) should be matched closely to minimize insertion losses, although this is not as critical in the case of sensors as it is in the case of optical communications devices. In the exemplary, illustrative, non-limiting implementation, the waveguides can be generally of either higher index of refraction than that of the fiber or physically smaller in at least one cross sectional dimension, or both. Antireflection coatings (ARC), formulated in a manner well known in the current art, can be applied to all mating optical surfaces. Both fibers and sensor waveguides can be tapered, and multiple layers of materials with differing indices of refraction (n) can be applied on top of the sensor waveguides (normal to the direction of light propagation), either completely or partially covering the waveguides.
The component of the magnetization vector out of the plane 14 can also cause a nonreciprocal change in the effective refractive index of the TM waveguide mode. However, the influence of the film component of magnetization perpendicular to the plane of the film is typically an order of magnitude or more less than that of the in-plane magnetization component, which is oriented perpendicular to the waveguide direction. The influence of the out-of-plane M component can be further minimized by the proper design of the waveguide structure, since the sensitivity of the nonreciprocal magneto optical effect depends on the geometry of the waveguide (roughly on the ratio of the thickness of the magneto-optical layer to the width of the rib waveguide). Growing iron garnet films with the proper compositions on a substrate of proper lattice constant and proper crystal orientation can also suppress the out-of plane magnetization component considerably.
According to one exemplary illustrative non-limiting implementation, a novel magnetic field sensor suitable for coupling to an optical fiber, and further embodying properties capable of inducing a shift in the wavelength position of an optical resonance, is illustrated schematically in
The structured layer 23 can be partially cut or etched into the MO film 22, by any method commonly known in the art, but in order to not interfere with the rotation of the magnetization vector, the structured layer 23 is preferably applied to the YIG surface and is of a non-magnetically active, transparent, lower index of refraction material. The grating can be formed in the layer 23 by ultraviolet photolithography, electron beam lithography, stamping or any other method known in the art.
The cladding layer 26 preferably possesses a lower index of refraction than that of the MO active material 22 or the waveguide material 23 and can be applied by sputtering, chemical vapor deposition (CVD) or any other method known in the art. Layer 26 can be comprised of multiple layers intended to protect the waveguide physically as well as to affect the characteristics of the reflected and transmitted light. Incident light from an optical fiber is partially reflected and partially transmitted, since the Bragg grating acts as a wavelength selective mirror. The degree of reflection, spectral position and spectral width of the reflected wavelength band depend, among other things, upon the length and period Λ of the alternating index grating segments, its width 27, its overall length and the index of refraction contrast between the alternating segments of the grating, and several other parameters known in the art.
The one or more phase shift elements 24, which can be incorporated to increase the resolution of the sensor, further causes a narrow, sharp valley in the reflected Bragg resonance peak. The end tapers of the waveguides 25, which may or may not be incorporated, aid in the NA matching between the fibers and the sensor waveguide. The tapered waveguide ends 24 are of length and taper designed to match as well as possible to the numerical aperture of the attached optical fiber, and can further have partial layers applied to their top surfaces, as is commonly done in the art, to further match the numerical apertures.
The waveguide 33 can be made of a high Verdet constant glass doped with an element or elements to increase the magneto-optical activity of the glass, for instance, with Tb. The waveguide can further be doped with one or materials to cause the glass to be sensitive to ultraviolet light, such as germanium, and to modify the index of refraction, such as boron and phosphorus and/or aluminum. The phase shift element 35, which optionally can be incorporated, causes a narrow, sharp valley in the reflected Bragg resonance peak that can be spectrally measured with higher resolution than a broad resonance peak. The end tapers of the waveguides 34, which may be incorporated, aid in the NA matching between the fibers and the sensor waveguide. The cladding 36 can be in general a lower refractive index material that physically protects the waveguide and further confines the waveguide optical modes to the waveguide. The waveguide may be formed in the high Verdet constant glass in several ways known in the art. If, as a non-limiting example, Ge is added to the glass, the grating can be formed by exposing the waveguide to a high intensity ultraviolet light through a phase mask, as is done with fiber Bragg gratings. The waveguide can also be formed with ultraviolet photolithography through a Cr-on-glass mask or by electron beam lithography and subsequently etched partially into or completely through the high Verdet constant glass using, for example, reactive ion etching
Since the effective refractive index of the waveguide mode depends on the magnetic field, the wavelength position of the exemplary valleys in the transmission will also depend on the value and direction of the applied magnetic field. Due to the nonreciprocity of the effect in the case of a perfectly circular ring, the sensitivity to the in-plane component of the external magnetic field will be very small if the external magnetic field is uniform across the ring area since the axially symmetric segments of the ring will cancel the effects of each other. However, such a resonant structure will be highly sensitive to the nonuniformity of the in-plane projection of the external magnetic field and thus can be used as a magnetic field gradient sensor. The effect of the component of magnetization perpendicular to the ring (uniform or not), however, will add up from all the segments of the ring and will thus be stronger than that from the in-plane, uniform components. Thus, waveguide ring (or disc) resonators can be used as sensors of uniform magnetic fields perpendicular to the plane of the substrate.
As a nonlimiting illustrative example,
While the technology herein has been described in connection with exemplary illustrative non-limiting implementations, the invention is not to be limited by the disclosure. The invention is intended to be defined by the claims and to cover all corresponding and equivalent arrangements whether or not specifically disclosed herein.
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7417425 *||Sep 5, 2006||Aug 26, 2008||International Superconductivity Technology Center,The Juridical Foundation||Continuous observation apparatus and method of magnetic flux distribution|
|US7421001 *||Jun 16, 2006||Sep 2, 2008||Pbc Lasers Gmbh||External cavity optoelectronic device|
|US7949031||Aug 28, 2008||May 24, 2011||Pbc Lasers Gmbh||Optoelectronic systems providing high-power high-brightness laser light based on field coupled arrays, bars and stacks of semicondutor diode lasers|
|US8009942 *||Jul 1, 2009||Aug 30, 2011||Duke University||Optical isolator|
|US8355419||Nov 30, 2009||Jan 15, 2013||Pbc Lasers Gmbh||Semiconductor optoelectronic device with improved beam quality|
|US8362762 *||Mar 16, 2010||Jan 29, 2013||Seiko Epson Corporation||Magnetic field measurement device|
|US8855451||Jul 20, 2012||Oct 7, 2014||Duke University||Optical isolator|
|US20100237858 *||Mar 16, 2010||Sep 23, 2010||Seiko Epson Corporation||Magnetic field measurement device|
|US20110311181 *||Dec 22, 2011||Duke University||Optical Isolator|
|EP2380029A1 *||Jan 16, 2009||Oct 26, 2011||Toyota Jidosha Kabushiki Kaisha||Ultrasensitive magnetic sensor based on giant faraday rotation|
|EP2380029A4 *||Jan 16, 2009||Sep 17, 2014||Toyota Motor Co Ltd||Ultrasensitive magnetic sensor based on giant faraday rotation|
|Jan 11, 2006||AS||Assignment|
Owner name: LAKE SHORE CRYOTRONICS, INC., OHIO
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KOCHERGIN, VLADIMI;SWINEHART, PHILIP;MAKLAD, MOKHTAR;REEL/FRAME:017451/0390;SIGNING DATES FROM 20051219 TO 20060109
|Jan 26, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Jan 27, 2015||FPAY||Fee payment|
Year of fee payment: 8
|Mar 3, 2015||AS||Assignment|
Owner name: KYTON, LLC, OHIO
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:LAKE SHORE CRYOTRONICS, INC.;REEL/FRAME:035077/0948
Effective date: 20150209